R: combining different X axes on a scatter plot when the Y axes are the same?

Question

I have a dataset that measures population index against temperature in warmer months and in colder months.

Instead of making a scatter plot that shows the summer temp against the pop index and then another showing the winter temp against pop index I'd like to combine the 2 - however the X axis in the winter dataset typically runs from -1 degrees to 10 degrees, and in the summer 10-25 degrees. Since they are the same scale is there any way I can combine the 2 X axes to have them next to eachother so summer and winter temps against population index can be shown in one scatter plot?

Right now I have plot(winter_RA, pop_RA) and plot(summer_RA, pop_RA) ; I tried plot(winter_RA+summer_RA, pop_RA) but it didn't come out showing the full range of temperatures on the X axis.

I'm completely new to R so sorry if this has an obvious answer. TIA

Answer 1

Since you have not provided the data along with your question, I'll answer it with the in-built dataset from RStudio called mtcars .

Here's how you can do it using ggplot .

mt <- ggplot(mtcars, aes(mpg, wt, colour = factor(cyl))) +
  geom_point()

mt + facet_grid(vars(cyl), scales = "free")

The plot will look like this:

As you can see in the code, you'll need to use facet_grid and provide vars ie variable. In your case, it will be seasons. And set scales = "free" .

Answer 2

Here is what I crafted. Not sure if it is close to what you are looking for, but I'm using the ggpubr package for these:

# Load ggpubr library for scatterplot:
library(ggpubr)

# Create temperature by population data frame:
cold_temp <- c(3,5,2,1,0,20)
hot_temp <- c(70,60,80,50,99,80)
pop <- c(500, 600, 200, 400, 300, 100)
temps <- data.frame(cold_temp,hot_temp, pop)

# Create scatterplot colored by temps:
ggscatter(temps,
       x="pop",
       y=c("hot_temp", "cold_temp"),
       merge = T)

Creates this plot:

Can decorate it more with this code:

# Create scatterplot colored by temps:
ggscatter(temps,
       x="pop",
       y=c("hot_temp", "cold_temp"),
       merge = T)+
  labs(title = "Average Temperature High/Cold by Population",
       subtitle = "Scatterplot Using GGPUBR Package",
       caption = "Data obtained from (insert place).",
       x="Population",
       y="Temperature (F)")+
  theme_bw()+
  theme(plot.title = element_text(face = "bold"),
        plot.caption = element_text(face = "italic"))

Which makes this:

As Vishal already noted, since there is no data present, it would be a bit easier, as you could potentially factor what is there. For example, you could use pivot_longer like so:

# Load tidyverse for "pivot_longer" function:
library(tidyverse)

# Pivot data:
pivot_temp <- temps %>% 
  pivot_longer(cols = c(hot_temp,cold_temp),
               names_to = "Temperature_Type",
               values_to = "Fahrenheit")

# Make faceted plot:
ggscatter(pivot_temp,
          x="pop",
          y="Fahrenheit",
          color = "Temperature_Type",
          palette = "jco")+
  facet_wrap(~Temperature_Type)+
  labs(title = "Average Temperature High/Cold by Population",
       subtitle = "Scatterplot Using GGPUBR Package",
       caption = "Data obtained from (insert place).",
       x="Population",
       y="Temperature (F)")+
  theme_bw()+
  theme(plot.title = element_text(face = "bold"),
        plot.caption = element_text(face = "italic"))

Which makes this:

Can alternatively add lines with:

  # Make faceted plot:
  ggscatter(pivot_temp,
            x="pop",
            y="Fahrenheit",
            color = "Temperature_Type",
            palette = "jco",
            merge = T)+
    geom_line(aes(color=Temperature_Type))+
    labs(title = "Average Temperature High/Cold by Population",
         subtitle = "Scatterplot Using GGPUBR Package",
         caption = "Data obtained from (insert place).",
         x="Population",
         y="Temperature (F)")+
    theme_bw()+
    theme(plot.title = element_text(face = "bold"),
          plot.caption = element_text(face = "italic"))

Which makes this slightly better looking line and scatterplot: